WO2022191173A1 - Gas identification method, and gas identification system - Google Patents
Gas identification method, and gas identification system Download PDFInfo
- Publication number
- WO2022191173A1 WO2022191173A1 PCT/JP2022/009944 JP2022009944W WO2022191173A1 WO 2022191173 A1 WO2022191173 A1 WO 2022191173A1 JP 2022009944 W JP2022009944 W JP 2022009944W WO 2022191173 A1 WO2022191173 A1 WO 2022191173A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- period
- value
- gas
- sensor
- signal
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000000605 extraction Methods 0.000 claims abstract description 86
- 238000001179 sorption measurement Methods 0.000 claims abstract description 23
- 238000005259 measurement Methods 0.000 claims description 140
- 239000000284 extract Substances 0.000 claims description 29
- 230000014509 gene expression Effects 0.000 claims description 12
- 239000007789 gas Substances 0.000 abstract description 277
- 239000000126 substance Substances 0.000 description 35
- 238000004891 communication Methods 0.000 description 29
- 239000013076 target substance Substances 0.000 description 20
- 238000010586 diagram Methods 0.000 description 18
- 239000000463 material Substances 0.000 description 16
- 101100425597 Solanum lycopersicum Tm-1 gene Proteins 0.000 description 15
- 238000001514 detection method Methods 0.000 description 15
- 230000008859 change Effects 0.000 description 13
- 229920005989 resin Polymers 0.000 description 9
- 239000011347 resin Substances 0.000 description 9
- 238000013528 artificial neural network Methods 0.000 description 7
- 238000010801 machine learning Methods 0.000 description 5
- 239000004065 semiconductor Substances 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 230000010354 integration Effects 0.000 description 4
- 239000012855 volatile organic compound Substances 0.000 description 4
- 238000004590 computer program Methods 0.000 description 3
- 239000000470 constituent Substances 0.000 description 3
- 238000013095 identification testing Methods 0.000 description 3
- LAQFLZHBVPULPL-UHFFFAOYSA-N methyl(phenyl)silicon Chemical compound C[Si]C1=CC=CC=C1 LAQFLZHBVPULPL-UHFFFAOYSA-N 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 125000001997 phenyl group Chemical group [H]C1=C([H])C([H])=C(*)C([H])=C1[H] 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- WRMNZCZEMHIOCP-UHFFFAOYSA-N 2-phenylethanol Chemical compound OCCC1=CC=CC=C1 WRMNZCZEMHIOCP-UHFFFAOYSA-N 0.000 description 2
- QGZKDVFQNNGYKY-UHFFFAOYSA-N Ammonia Chemical compound N QGZKDVFQNNGYKY-UHFFFAOYSA-N 0.000 description 2
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- 230000005526 G1 to G0 transition Effects 0.000 description 2
- 230000004308 accommodation Effects 0.000 description 2
- 239000012491 analyte Substances 0.000 description 2
- 239000012159 carrier gas Substances 0.000 description 2
- 238000007599 discharging Methods 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 229920002050 silicone resin Polymers 0.000 description 2
- ZFRKQXVRDFCRJG-UHFFFAOYSA-N skatole Chemical compound C1=CC=C2C(C)=CNC2=C1 ZFRKQXVRDFCRJG-UHFFFAOYSA-N 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- MVOSYKNQRRHGKX-UHFFFAOYSA-N 11-Undecanolactone Chemical compound O=C1CCCCCCCCCCO1 MVOSYKNQRRHGKX-UHFFFAOYSA-N 0.000 description 1
- XYHKNCXZYYTLRG-UHFFFAOYSA-N 1h-imidazole-2-carbaldehyde Chemical compound O=CC1=NC=CN1 XYHKNCXZYYTLRG-UHFFFAOYSA-N 0.000 description 1
- CFAKWWQIUFSQFU-UHFFFAOYSA-N 2-hydroxy-3-methylcyclopent-2-en-1-one Chemical compound CC1=C(O)C(=O)CC1 CFAKWWQIUFSQFU-UHFFFAOYSA-N 0.000 description 1
- GWYFCOCPABKNJV-UHFFFAOYSA-M 3-Methylbutanoic acid Natural products CC(C)CC([O-])=O GWYFCOCPABKNJV-UHFFFAOYSA-M 0.000 description 1
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 1
- 239000003463 adsorbent Substances 0.000 description 1
- 229910021529 ammonia Inorganic materials 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- GWYFCOCPABKNJV-UHFFFAOYSA-N beta-methyl-butyric acid Natural products CC(C)CC(O)=O GWYFCOCPABKNJV-UHFFFAOYSA-N 0.000 description 1
- 229910002091 carbon monoxide Inorganic materials 0.000 description 1
- 239000000919 ceramic Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 239000013078 crystal Substances 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000005669 field effect Effects 0.000 description 1
- 239000010419 fine particle Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000004817 gas chromatography Methods 0.000 description 1
- 239000011261 inert gas Substances 0.000 description 1
- 229910001872 inorganic gas Inorganic materials 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 125000002496 methyl group Chemical group [H]C([H])([H])* 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 229940067107 phenylethyl alcohol Drugs 0.000 description 1
- 229920001515 polyalkylene glycol Polymers 0.000 description 1
- 229920001225 polyester resin Polymers 0.000 description 1
- 239000004645 polyester resin Substances 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 229940074386 skatole Drugs 0.000 description 1
- 125000001424 substituent group Chemical group 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000010897 surface acoustic wave method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/0004—Gaseous mixtures, e.g. polluted air
- G01N33/0009—General constructional details of gas analysers, e.g. portable test equipment
- G01N33/0027—General constructional details of gas analysers, e.g. portable test equipment concerning the detector
- G01N33/0031—General constructional details of gas analysers, e.g. portable test equipment concerning the detector comprising two or more sensors, e.g. a sensor array
- G01N33/0034—General constructional details of gas analysers, e.g. portable test equipment concerning the detector comprising two or more sensors, e.g. a sensor array comprising neural networks or related mathematical techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
- G01N27/02—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance
- G01N27/04—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance by investigating resistance
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
- G01N27/02—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance
- G01N27/04—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance by investigating resistance
- G01N27/12—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance by investigating resistance of a solid body in dependence upon absorption of a fluid; of a solid body in dependence upon reaction with a fluid, for detecting components in the fluid
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N5/00—Analysing materials by weighing, e.g. weighing small particles separated from a gas or liquid
- G01N5/02—Analysing materials by weighing, e.g. weighing small particles separated from a gas or liquid by absorbing or adsorbing components of a material and determining change of weight of the adsorbent, e.g. determining moisture content
Definitions
- the present disclosure relates to gas identification methods and gas identification systems.
- Patent Document 1 discloses a method of identifying an analyte using data of a pulsed signal that detects the analyte, using the intensity, wavelength, intensity ratio, kurtosis, etc. of the pulsed signal as feature quantities.
- the present disclosure provides a gas identification method and the like that can improve identification accuracy.
- a gas identification method is a gas identification method using a sensor that outputs a signal corresponding to the adsorption concentration of a gas, comprising a first period, a second period following the first period, and a first step of acquiring a signal output from the sensor exposed to the sample gas only during the second period of a measurement period consisting of a third period following the second period; and a drift of the acquired signal.
- a gas identification system includes a sensor that outputs a signal corresponding to the adsorption concentration of a gas, a first period, a second period following the first period, and a third period following the second period.
- an exposure unit that exposes the sensor to the sample gas only during the second period of the measurement period consisting of periods; an acquisition circuit that acquires the signal output from the sensor during the measurement period; and a drift of the acquired signal.
- an extraction circuit for extracting one or more feature quantities corresponding to the above, a memory storing a learned logical model for identifying the sample gas, and the one or more extracted feature quantities using the learned logical model an identification circuit that identifies the sample gas based on the identification information and outputs an identification result.
- identification accuracy can be improved.
- FIG. 1 is a block diagram showing a schematic configuration of a gas identification system according to an embodiment.
- FIG. 2 is a schematic diagram showing an example of a configuration of an exposed portion according to the embodiment.
- FIG. 3 is a block diagram showing a schematic configuration of a gas identification system according to a modification of the embodiment.
- FIG. 4 is a flow chart for explaining the operation of the gas identification system according to the embodiment.
- FIG. 5 is a diagram illustrating an example of a control signal to an exposure unit and a signal output from a sensor according to the embodiment;
- FIG. 6 is a diagram showing an example of signals output from the sensor according to the embodiment in a plurality of continuous measurement periods;
- FIG. 7 is a diagram for explaining values of signals acquired by the extraction circuit according to the embodiment.
- FIG. 1 is a block diagram showing a schematic configuration of a gas identification system according to an embodiment.
- FIG. 2 is a schematic diagram showing an example of a configuration of an exposed portion according to the embodiment.
- FIG. 3 is
- FIG. 8 is another diagram for explaining values of signals acquired by the extraction circuit according to the embodiment.
- FIG. 9 is a diagram for explaining values of signals acquired by the extraction circuit according to the embodiment in a plurality of continuous measurement periods;
- FIG. 10 is another diagram for explaining signal values acquired by the extraction circuit according to the embodiment in a plurality of consecutive measurement periods.
- 11A and 11B are diagrams for explaining the rate and amount of change in the value of the signal output from the sensor according to the embodiment.
- a sensor that outputs a signal corresponding to the adsorption concentration of a gas is used for gas identification, for example, a feature value using the signal output from the sensor when exposed to a sample gas containing chemical substances such as volatile organic compounds
- the chemical substance contained in the sample gas is identified as the substance to be identified.
- the signal output from the sensor changes because the concentration of adsorption to the sensor differs depending on the type of chemical substance. Therefore, chemical substances contained in the sample gas are identified using, for example, the amount of change and the rate of change in the signal output from the sensor during the period in which the sensor is exposed to the sample gas as the feature amount.
- the gas identification method is required to improve the identification accuracy.
- the present disclosure provides a gas identification method and the like that can improve identification accuracy based on such knowledge.
- a gas identification method is a gas identification method using a sensor that outputs a signal corresponding to the adsorption concentration of a gas, comprising a first period, a second period following the first period, and a first step of acquiring a signal output from the sensor exposed to the sample gas only during the second period of a measurement period consisting of a third period following the second period; and a drift of the acquired signal.
- the drift of the signal which is different from the output of the signal during the second period in which the sensor is exposed to the sample gas, that is, the output depending on the adsorption concentration of the gas to the sensor, is determined.
- One or more corresponding feature quantities are extracted. Therefore, in the third step, identification based on signal drift is performed, and even when sample gases with similar outputs from sensors that depend on gas adsorption concentration are identified, sample gases can be identified with high identification accuracy.
- one or more feature values can be extracted using the signal output from the sensor exposed in the fourth step.
- the senor may be exposed to a reference gas during the first period and the third period.
- the signal output from the sensor may be acquired via a network.
- the value of the signal fluctuated due to the exposure of the sensor to the sample gas during the second period is about to return to a reference value during the third period.
- the value of the signal fluctuated due to the exposure of the sensor to the sample gas during the second period is about to return to a reference value during the third period.
- the first value which is the value of the signal when trying to return to the reference value directly related to drift, can be used to extract the feature amount.
- the signal output from the sensor is acquired in a plurality of consecutive measurement periods
- two or more of the measurement periods out of the plurality of measurement periods are acquired.
- the first values may be obtained for each, and a difference between the obtained first values may be extracted as at least one feature quantity among the one or more feature quantities.
- the signal output from the sensor is acquired in a plurality of consecutive measurement periods
- two or more of the measurement periods out of the plurality of measurement periods are acquired.
- Obtaining the first value for each deriving an approximate expression using the obtained first value, and extracting a coefficient of the derived approximate expression as at least one feature quantity among the one or more feature quantities.
- the feature amount in which the variation in the first value is smoothed out can be extracted.
- a second value that is the value of the last signal in the first period is acquired, and the difference between the acquired first value and the acquired second value is calculated as the 1 At least one of the above feature amounts may be extracted.
- the feature quantity corresponding to the drift can be extracted using the second value as the reference value.
- the value of the signal at the end of the third period may be obtained as the first value.
- the value of the signal at the end of the third period in which the value of the signal that is about to return to the reference value in the third period tends to be stable As a result, it is possible to use, as the first value, the value of the signal at the end of the third period in which the value of the signal that is about to return to the reference value in the third period tends to be stable.
- the signal output from the sensor is acquired in a plurality of consecutive measurement periods
- two or more of the measurement periods out of the plurality of measurement periods are acquired.
- a third value which is the value of the signal when the sensor fluctuates due to exposure to the sample gas in the second period, is obtained, and the difference between the obtained third values is calculated as the At least one of the one or more feature amounts may be extracted.
- the signal output from the sensor is acquired in a plurality of consecutive measurement periods
- two or more of the measurement periods out of the plurality of measurement periods are acquired.
- a third value which is the value of the signal when the sensor is exposed to the sample gas during the second period
- an approximate expression is derived using the obtained third value.
- the derived coefficient of the approximate expression may be extracted as at least one of the one or more feature amounts.
- the signal output from the sensor is obtained in a plurality of consecutive measurement periods
- the second step in the second and subsequent measurement periods of the plurality of measurement periods At least one of the one or more feature amounts may be extracted based on the acquired signal.
- the feature quantity is extracted using the signals output in the second and subsequent measurement periods in which the drift is likely to increase. Therefore, the difference in the extracted feature amount tends to increase according to the type of sample gas, and the identification accuracy can be further improved.
- a gas identification system includes a sensor that outputs a signal corresponding to the adsorption concentration of a gas, a first period, a second period following the first period, and a period following the second period.
- an exposure unit that exposes the sensor to the sample gas only during the second period of a measurement period consisting of a third period; an acquisition circuit that acquires a signal output from the sensor during the measurement period; and the acquired signal.
- an extraction circuit for extracting one or more feature quantities corresponding to the drift of the sample gas, a memory in which a learned logical model for identifying the sample gas is stored, and the one or more features extracted using the learned logical model an identification circuit for identifying the sample gas based on the quantity and outputting an identification result.
- the extraction circuit responds to the drift of the signal, which is different from the output of the signal during the second period in which the sensor is exposed to the sample gas, that is, the output dependent on the adsorption concentration of the gas to the sensor. 1 or more feature quantities are extracted. Therefore, the identification circuit performs identification based on signal drift, and can identify sample gases with high identification accuracy even when identifying sample gases with similar outputs from sensors that depend on gas adsorption concentrations.
- each figure is not necessarily a strict illustration.
- substantially the same configurations are denoted by the same reference numerals, and overlapping descriptions are omitted or simplified.
- ordinal numbers such as “first” and “second” do not mean the number or order of steps, components, etc., unless otherwise specified. It is used for the purpose of avoiding confusion and distinguishing between
- FIG. 1 is a block diagram showing a schematic configuration of a gas identification system 100 according to this embodiment.
- a gas identification system 100 includes a sensor 10, an exposure unit 20, a control circuit 31, an acquisition circuit 32, an extraction circuit 33, an identification circuit 34, a memory 40 and.
- Gas identification system 100 identifies a sample gas based on the output of sensor 10 exposed to the sample gas.
- the sample gas contains, for example, chemical substances to be identified.
- the sample gas is, for example, gas collected from food, exhaled air collected from the human body, air surrounding the human body, air collected from a room in a building, or the like.
- the gas identification system 100 identifies chemical substances contained in the sample gas, for example. Specifically, the gas identification system 100 identifies which of the plurality of identification target substances is contained in the sample gas as a chemical substance. Further, the gas identification system 100 may identify whether or not the sample gas contains the identification target substance.
- Substances to be identified are, for example, volatile organic compounds, but may also be inorganic gases such as ammonia and carbon monoxide.
- the gas identification system 100 is used, for example, for odor identification.
- the volatile organic compound is, for example, a molecule that becomes an odor component.
- the sensor 10 is a sensor that outputs a signal corresponding to the adsorption concentration of gas.
- the sensor 10 is, for example, an electrochemical sensor, a semiconductor sensor, a field effect transistor sensor, a surface acoustic wave sensor, a crystal oscillator sensor, or a variable resistance sensor.
- the sensor 10 has, for example, a sensing portion and a pair of electrodes electrically connected to the sensing portion.
- the sensing part changes its electric resistance value according to, for example, the adsorption concentration of the gas.
- a signal corresponding to the electrical resistance value of the sensing portion of the sensor 10 is acquired by the acquisition circuit 32 as, for example, a voltage signal or a current signal via a pair of electrodes.
- the sensing part is composed of, for example, a resin material that is an adsorbent that adsorbs gas and conductive particles dispersed in the resin material.
- resin materials include polyalkylene glycol resins, polyester resins and silicone resins.
- the resin material is, for example, a material commercially available as a stationary phase of a gas chromatography column in the side chain. From the viewpoint of durability and gas adsorption, the resin material may be, for example, a silicone resin having various substituents such as phenyl groups and methyl groups on the side chains, which is commercially available as a stationary phase for columns.
- the sensing portion is not limited to the configuration of the resin material and the conductive particles, and may be any member as long as the electrical resistance value changes due to the adsorption of gas. It may be made of porous ceramics, for example.
- the gas identification system 100 includes a plurality of sensors 10, for example. At least two sensors 10 among the plurality of sensors 10 each have a sensing portion (specifically, a resin material that constitutes the sensing portion), which is made of, for example, different kinds of materials. Also, the types of materials of the sensing portions of all the plurality of sensors 10 may be different from each other. Different types of materials exhibit different adsorption behaviors with respect to the same chemical substance. Therefore, the multiple sensors 10 output different signals for the same chemical substance. As a result, different feature quantities can be extracted from the outputs of the plurality of sensors 10, so that the identification accuracy of the gas identification system 100 can be improved.
- a sensing portion specifically, a resin material that constitutes the sensing portion
- the types of materials of the sensing portions of all the plurality of sensors 10 may be different from each other. Different types of materials exhibit different adsorption behaviors with respect to the same chemical substance. Therefore, the multiple sensors 10 output different signals for the same chemical substance. As a result, different feature quantities can be extracted from the
- the exposure unit 20 is an exposure mechanism that exposes the sensor 10 to gas under the control of the control circuit 31 . Specifically, the exposure unit 20 exposes the sensor 10 to the sample gas only during the second period of the measurement period consisting of the first period, the second period following the first period, and the third period following the second period. expose. Also, the exposure unit 20 may expose the sensor 10 to the reference gas during the first period and the third period.
- a reference gas is a gas that serves as a reference for measurement, and is, for example, a gas that does not contain a substance to be identified. Also, the reference gas is, for example, a gas that is less likely to be adsorbed by the sensing part of the sensor 10 than the substance to be identified.
- the reference gas examples include air, an inert gas such as nitrogen, and a gas obtained by removing chemical substances from a sample gas using a filter or the like.
- FIG. 2 is a schematic diagram showing an example of the configuration of the exposed portion 20 according to this embodiment.
- the exposure section 20 has, for example, a housing section 21, a three-way solenoid valve 22, an intake pump 23, and a plurality of pipes 25a, 25b, 25c, 25d, and 25e.
- An inlet port 26a for introducing sample gas is provided at one end of the pipe 25a.
- the intake port 26a is provided, for example, in a space filled with sample gas.
- One end of the pipe 25b is provided with an intake port 26b for introducing the reference gas.
- the intake port 26b is provided, for example, in a space filled with a reference gas.
- One end of the pipe 25e is provided with an exhaust port 26e for discharging the introduced sample gas and reference gas.
- the housing portion 21 is a box-shaped container that houses the sensor 10 .
- a plurality of sensors 10 are arranged in an array inside the housing portion 21 .
- One end of each of the pipe 25c and the pipe 25d is connected to the accommodation portion 21 .
- a plurality of sensors 10 are arranged in a gas flow path.
- the sample gas introduced from the intake port 26a is introduced into the housing portion 21 via the pipe 25a, the three-way solenoid valve 22, and the pipe 25c.
- the reference gas introduced from the intake port 26b is introduced into the housing portion 21 via the pipe 25b, the three-way solenoid valve 22, and the pipe 25c.
- the sample gas and the reference gas introduced into the storage section 21 are discharged from the exhaust port 26e via the pipe 25d, the intake pump 23 and the pipe 25e.
- the three-way solenoid valve 22 is a solenoid valve for switching the gas to be introduced into the housing portion 21 .
- the three-way solenoid valve 22 has an input port P1 to which the other end of the pipe 25a is connected, an input port P2 to which the other end of the pipe 25b is connected, and an output port P3 to which the other end of the pipe 25c is connected. is provided.
- the three-way solenoid valve 22 is controlled by the control circuit 31 to open and close each port. Under the control of the control circuit 31, the three-way solenoid valve 22 switches between a first state in which the input port P1 and the output port P3 are electrically connected and a second state in which the input port P2 and the output port P3 are electrically connected. In a first state, input port P1 and output port P3 are open and input port P2 is closed. Also, in the second state, the input port P2 and the output port P3 are open, and the input port P1 is closed.
- the intake pump 23 is a pump for introducing the sample gas and the reference gas into the housing section 21 and discharging the introduced sample gas and reference gas from the exhaust port 26e.
- the operation of the intake pump 23 is controlled by the control of the control circuit 31 .
- An intake port of the intake pump 23 is connected to the other end of the pipe 25d.
- the exhaust port of the intake pump 23 is connected to the other end of the pipe 25e.
- the sample gas is introduced into the housing portion 21 .
- the exposure unit 20 exposes the plurality of sensors 10 to the sample gas.
- the reference gas is introduced into the housing portion 21 .
- the exposure unit 20 exposes the plurality of sensors 10 to the reference gas.
- the configuration of the exposure unit 20 is not limited to the configuration shown in FIG. 2, and is not particularly limited as long as the configuration allows the sensor 10 to be exposed to the sample gas.
- the exposure unit 20 may be configured, for example, such that the sample gas and the reference gas are introduced into the storage unit 21 through separate pipes without passing through the three-way solenoid valve 22 .
- the exposure section 20 may have a configuration in which the sample gas is mixed into the carrier gas without the suction pump 23, and the carrier gas is always flowed to the storage section 21.
- the intake pump 23 may be used to evacuate the storage section 21 .
- the exposure unit 20 further includes various removal filters for removing moisture or fine particles from the sample gas and reference gas, an electromagnetic control valve for adjusting the flow rate of each pipe, and a check valve for preventing reverse flow in each pipe. may
- control circuit 31 controls the operation of the exposing section 20, specifically the three-way solenoid valve 22 and the intake pump 23, as described above.
- the control circuit 31 may also output information indicating the timing of the operation of the exposing section 20 to the acquisition circuit 32 .
- the acquisition circuit 32 acquires the signal output from the sensor 10 during the measurement period.
- the acquisition circuit 32 acquires, for example, a voltage signal or a current signal as a signal output corresponding to the electrical resistance value of the sensing portion of the sensor 10 .
- the extraction circuit 33 extracts one or more feature quantities corresponding to the drift of the signal acquired by the acquisition circuit 32 .
- the extraction circuit 33 may extract a feature quantity other than the feature quantity corresponding to the drift from the signal acquired by the acquisition circuit 32 .
- the extraction circuit 33 extracts one or more feature quantities from the signals output by each of the multiple sensors 10 .
- the identification circuit 34 identifies the sample gas based on one or more feature quantities extracted by the extraction circuit 33 using the learned logical model.
- the identification circuit 34 identifies, for example, which of the plurality of identification target substances is contained in the sample gas. Further, the identification circuit 34 may identify whether or not the sample gas contains the identification target substance.
- the identification circuit 34 receives one or more feature values as input and outputs identification results.
- the identification circuit 34 outputs, for example, information for displaying the identification result on a display (not shown) or the like provided in the gas identification system.
- the identification circuit 34 may output information indicating the identification result to the memory 40 and cause the memory 40 to store the information. Further, the identification circuit 34 may output information indicating the identification result to an external device.
- the control circuit 31, the acquisition circuit 32, the extraction circuit 33, and the identification circuit 34 are implemented by a microcomputer or processor containing a program that performs the above processes.
- the control circuit 31, the acquisition circuit 32, the extraction circuit 33, and the identification circuit 34 may each be implemented by a dedicated logic circuit that performs the above processing.
- the memory 40 is a storage device that stores the learned logical model used in the identification circuit 34.
- the memory 40 is implemented by, for example, a semiconductor memory.
- a learned logical model is a logical model that identifies a sample gas.
- the learned logical model is, for example, a logical model that identifies which of the plurality of identification target substances is contained in the sample gas.
- the learned logic model receives, for example, one or more feature values extracted by the extraction circuit 33 as input, and outputs which of the plurality of identification target substances is contained in the sample gas.
- the learned logic model may output whether or not the sample gas contains the substance to be identified.
- the learned logical model is constructed by performing machine learning using, for example, a known identification target substance and one or more feature values extracted by the extraction circuit 33 using the known identification target substance as teacher data.
- the method used to construct the logical model in machine learning is not particularly limited.
- a neural network for example, is used to construct a logical model in machine learning. That is, a trained logical model includes, for example, a neural network.
- a random forest, a support vector machine, a self-organizing map, or the like may be used to construct a logical model in machine learning.
- the gas identification system 100 is implemented as, for example, a single gas identification device including the above components, but may be implemented by a plurality of devices. When the gas identification system 100 is implemented by multiple devices, the components included in the gas identification system 100 may be distributed among the multiple devices in any way.
- FIG. 3 is a block diagram showing a schematic configuration of a gas identification system 100a according to a modification of the embodiment.
- the gas identification system 100a includes a detection device 200 and an identification device 300.
- the detection device 200 includes a sensor 10 , an exposure section 20 , a control circuit 31 , a detection section 50 and a communication section 51 .
- the sensor 10, the exposure unit 20 and the control circuit 31 have, for example, the same configuration as the gas identification system 100 described above.
- the detection unit 50 acquires the signal output from the sensor 10 during the measurement period. For example, a voltage signal or a current signal is acquired as a signal corresponding to the electrical resistance value of the sensing portion of the sensor 10 . Also, information indicating the timing of controlling the exposure unit 20 is acquired from the control circuit 31 . The detection unit 50 transmits the acquired signal and information to the identification device 300 using the communication unit 51 .
- the detection unit 50 is implemented by a microcomputer or processor containing a program for performing the above processes.
- the detection unit 50 may be implemented by a dedicated logic circuit that performs the above processing.
- the communication unit 51 is a communication module (communication circuit) for the detection device 200 to communicate with the identification device 300 via a wide area communication network 90 such as the Internet, which is an example of a network.
- the communication unit 51 may perform wired communication or wireless communication.
- a communication standard used for communication performed by the communication unit 51 is not particularly limited.
- the identification device 300 includes an acquisition circuit 32a, an extraction circuit 33, an identification circuit 34, a memory 40, and a communication unit 60.
- the extraction circuit 33, the identification circuit 34 and the memory 40 have, for example, the same configuration as the gas identification system 100 described above.
- the acquisition circuit 32a acquires the signal output from the sensor 10 during the measurement period, which is acquired by the detection unit 50 via the wide area communication network 90. Acquisition circuit 32 a communicates with detection device 200 via wide area communication network 90 using communication unit 60 .
- the communication unit 60 is a communication module (communication circuit) for the identification device 300 to communicate with the detection device 200 via the wide area communication network 90 .
- the communication unit 60 may perform wired communication or wireless communication.
- a communication standard used for communication performed by the communication unit 60 is not particularly limited.
- FIG. 4 is a flowchart for explaining the operation of the gas identification system 100 according to this embodiment.
- FIG. 4 is a flowchart of the gas identification method performed by gas identification system 100 .
- a gas identification method according to the present embodiment includes an exposure step, an acquisition step, an extraction step, and an identification step.
- the exposure step is an example of a fourth step
- the acquisition step is an example of a first step
- the extraction step is an example of a second step
- the identification step is an example of a third step.
- step S11 the exposure unit 20 exposes the sensor 10 to the sample gas only during the second period of the measurement period. Also, the exposure unit 20 exposes the sensor 10 to the reference gas during the first period and the third period.
- the control circuit 31 operates the intake pump 23 and controls opening and closing of each port of the three-way solenoid valve 22 to expose the sensor 10 to the reference gas during the first period and the third period, and expose the sensor 10 to the reference gas during the second period. exposing the sensor 10 to the sample gas.
- step S11 for example, in a plurality of continuous measurement periods, the sensor 10 is exposed to the sample gas only during the second period of the measurement periods.
- the acquisition circuit 32 acquires the signal output from the sensor 10 exposed in step S11 (step S12). That is, the acquisition circuit 32 acquires the signal output from the sensor 10 exposed to the sample gas only during the second period of the measurement period.
- the detector 50 acquires the signal output from the sensor 10 exposed in step S11.
- the acquisition circuit 32a acquires the signal output from the sensor 10 exposed in step S11 from the detection unit 50 via the wide area communication network 90. FIG. Accordingly, even when the sensor 10 is located at a location distant from the identification device 300, the acquisition circuit 32a can acquire the signal output from the sensor 10. FIG.
- FIG. 5 is a diagram showing an example of the control signal to the exposure unit 20 and the signal output from the sensor 10.
- FIG. FIG. 5 shows the intensity of the control signal to the three-way solenoid valve 22 and the signal output from the sensor 10 during the measurement period Tm consisting of the first period T1, the second period T2 and the third period T3.
- (a) of FIG. 5 is a graph showing an example of the change over time of the control signal output from the control circuit 31 .
- the control signal when the control signal is at High level, the three-way solenoid valve 22 is controlled to be in the first state, and when the control signal is at Low level, the three-way solenoid valve 22 is It is controlled to be in the second state.
- (b) of FIG. 5 is a graph showing an example of temporal changes in the intensity (for example, voltage) of the signal output from the sensor 10 .
- step S11 for example, as shown in (a) of FIG. 5, the three-way solenoid valve 22 is in the second state during the first period T1 and the third period T3, and the exposing section 20 exposes the sensor 10 to the reference gas. exposed to Also, during the second period T2, the three-way solenoid valve 22 is in the first state, and the exposure section 20 exposes the sensor 10 to the sample gas.
- the signal acquired in step S12 changes, for example, as shown in FIG. 5(b). First, during the first period T1 during which the sensor 10 is exposed to the reference gas, the signal value hardly changes.
- the sensing portion of the sensor 10 absorbs the sample gas (mainly chemical substances contained in the sample gas), and the signal value changes (for example, Rise.
- the sample gas mainly the chemical substances contained in the sample gas
- the signal value changes (for example, Rise.
- the sample gas mainly the chemical substances contained in the sample gas
- the value of the signal tries to return to the reference value.
- the reference value is, for example, the value before the sensor 10 is exposed to the sample gas and the signal value begins to fluctuate.
- the lengths of the first period, the second period, and the third period are not particularly limited, and are set according to, for example, the type of sensor 10 and the type of substance to be identified.
- the length of the first period T1 is, for example, 1 second or more and 10 seconds or less.
- the length of the second period T2 is, for example, 5 seconds or more and 30 seconds or less.
- the length of the third period is, for example, 10 seconds or more and 100 seconds or less.
- step S11 for example, the exposure unit 20 exposes the sensor 10 to the sample gas only during the second period of each measurement period Tm in a plurality of consecutive measurement periods Tm. Then, in step S12, the acquisition circuit 32 acquires signals output from the sensor 10 during a plurality of continuous measurement periods Tm.
- FIG. 6 is a diagram showing an example of signals output from the sensor 10 during a plurality of continuous measurement periods.
- the acquisition circuit 32 acquires the signal output from the sensor 10, for example, during seven consecutive measurement periods Tm-1 to Tm-7.
- the operation described with reference to FIG. 5 is performed, and the same operation is repeatedly performed.
- the next measurement period Tm starts before the value of the signal that fluctuated in the second period T2 completely returns to the reference value. drift increases.
- the number of consecutive measurement periods Tm is not particularly limited, and is set according to, for example, the type of sensor 10 and the type of substance to be identified.
- the extraction circuit 33 extracts one or more feature amounts corresponding to the drift of the signal acquired in step S12 (step S13).
- the extraction circuit 33 extracts one or more feature quantities using the acquired signal values.
- the baseline is dissociated from the reference value and drift occurs.
- a signal drift is considered to be caused, for example, by the difficulty of gas escape in the sensing portion of the sensor 10.
- the size changes depending on the combination with the material. For example, when the sample gas is difficult to escape from the sensing portion of the sensor 10, the value of the output from the sensor 10 does not return to the reference value and the drift increases. On the other hand, when the sample gas easily escapes from the sensing portion of the sensor 10, the drift is small or does not occur.
- the extraction circuit 33 extracts one or more feature quantities corresponding to signal drift for use in identifying the sample gas.
- the extraction circuit 33 obtains at least one of a first value, a second value, and a third value as the value of the signal from the obtained signal, and uses the obtained value to extract one or more feature amounts. do.
- FIGS. 7 and 8 are diagrams for explaining values of signals acquired by the extraction circuit 33.
- FIG. FIGS. 7 and 8 show changes over time in the intensity (for example, voltage) of the signal output from the sensor 10 acquired by the acquisition circuit 32.
- FIG. 7 and 8 show changes over time in the intensity (for example, voltage) of the signal output from the sensor 10 acquired by the acquisition circuit 32.
- the first value is the signal value when the signal value fluctuated due to the exposure of the sensor 10 to the sample gas during the second period T2 is about to return to the reference value during the third period T3.
- the first value is, for example, the signal value V1 at a predetermined point in the third period T3, as shown in FIG.
- the first value may be the average value of the signal values in the predetermined section S1 in the third period T3, as shown in FIG.
- the length of the section S1 is, for example, 0.1 seconds or more and 5 seconds or less.
- the first value may be the last signal value in the third period T3.
- the first value may be the signal value V1 at the last point in the third period T3, or the average value of the signal values in the interval S1 including the last point in the third period T3. may This allows the extraction circuit 33 to use the value of the signal that is stable during the third period T3 as the first value.
- the first value may be the value V1 of the signal at the time when a predetermined time has passed since the start of the third period T3, or the signal in the section S1 that starts after the predetermined time has passed since the start of the third period T3. It may be the average value of the values of The predetermined time is, for example, a time longer than half the third period T3.
- the first value is the value of the signal at the third time period T3 during which the sensor 10 is not exposed to the sample gas after the second time period T2 during which the sensor 10 is exposed to the sample gas. is less sensitive to exposure to , and is a good indicator of baseline fluctuations in the signal.
- the second value is the last signal value in the first period T1.
- the second value is, for example, the value V2 of the signal at the last point in the first period T1, as shown in FIG.
- the second value may be the average value of the signal values in the section S2 including the end of the first period T1, as shown in FIG.
- the length of the section S2 is, for example, 0.1 seconds or more and 5 seconds or less.
- the second value is the value of the signal during the first time period T1 when the sensor 10 is not exposed to the sample gas, prior to the second time period T2 when the sensor 10 is exposed to the sample gas. It is suitable as a value indicating a reference value.
- the third value is the value of the signal when it fluctuates due to the exposure of the sensor 10 to the sample gas during the second period T2.
- the third value is, for example, the signal value V3 at a predetermined point in the second period T2, as shown in FIG.
- the third value may be the average value of the signal values in the predetermined section S3 in the second period T2, as shown in FIG.
- the length of the section S3 is, for example, 0.1 seconds or more and 5 seconds or less.
- the third value is, for example, the value of the signal at the timing when the value of the signal becomes maximum in the second period T2.
- the third value may be the signal value V3 at the point in time when the signal value reaches its maximum value in the second period T2.
- the third value may be the value V3 of the signal at the time when a predetermined time has passed since the start of the second period T2, or the signal in the section S3 that starts after the predetermined time has passed since the start of the second period T2. It may be the average value of the values of The predetermined time is, for example, a time longer than half the second period T2.
- the third value is a value when it fluctuates due to adsorption of the sample gas on the sensor 10, and when the baseline of the signal shifts, the third value also shifts. Therefore, it is possible to extract the feature amount corresponding to the drift using the third value.
- the extraction circuit 33 extracts at least one of a first value, a second value, and a third value from the signal output from the sensor 10 in each of two or more measurement periods among a plurality of consecutive measurement periods Tm1 to Tm7. get one. Acquiring signals in a plurality of continuous measurement periods Tm1 to Tm7 in this way tends to increase the drift of the acquired signals, and can improve the identification accuracy described later.
- 9 and 10 are diagrams for explaining the values of the signals acquired by the extraction circuit 33 during a plurality of continuous measurement periods Tm-1 to Tm-7. 9 and 10 show temporal changes in the intensity (for example, voltage) of the signal output from the sensor 10 acquired by the acquisition circuit 32.
- FIG. 9 shows temporal changes in the intensity (for example, voltage) of the signal output from the sensor 10 acquired by the acquisition circuit 32.
- the extraction circuit 33 when obtaining the first value, obtains at least one of the first values in each of the first measurement period Tm-1 to the seventh measurement period Tm-7. get one. Acquisition of the second and third values is similar to acquisition of the first value.
- the extraction circuit 33 acquires the first value, the second value, or the third value in a plurality of measurement periods Tm, for example, in each measurement period Tm, the first value, the second value, or the third value is obtained at the same timing. get.
- the extraction circuit 33 acquires, for example, at least one of the signal values V1-1 to V1-7 corresponding to the signal value V1 shown in FIG. 7 as the first value.
- the extraction circuit 33 obtains, for example, the average value of the signal values in at least one of the intervals S1-1 to S1-7 corresponding to the interval S1 shown in FIG. 8 as the first value.
- the extraction circuit 33 acquires, for example, at least one of the signal values V3-1 to V3-7 corresponding to the signal value V3 shown in FIG. 7 as the third value.
- the extraction circuit 33 obtains, for example, the average value of the signal values in at least one of the intervals S3-1 to S3-7 corresponding to the interval S3 shown in FIG. 8 as the third value.
- obtaining the second value is similar to the first and third values.
- the extraction circuit 33 acquires the first value in each of two or more measurement periods Tm among a plurality of continuous measurement periods Tm-1 to Tm-7 shown in FIGS. 9 and 10, for example. , the difference between the acquired first values is extracted as a feature amount.
- the extraction circuit 33 extracts, for example, the difference between the first values of the two separated measurement periods Tm as the feature amount.
- the two spaced apart measurement periods Tm are, for example, a measurement period Tm-2 and a measurement period Tm-7, which are the second and final measurement periods Tm among a plurality of consecutive measurement periods Tm-1 to Tm-7.
- the extraction circuit 33 may also extract the difference between the first values of two consecutive measurement periods Tm (for example, the measurement period Tm-2 and the measurement period Tm-3) as the feature amount.
- the extraction circuit 33 may extract a plurality of feature quantities by changing a plurality of combinations of the two measurement periods Tm for obtaining the difference of the first values.
- the extraction circuit 33 may, for example, extract differences between the first values in all combinations of two consecutive measurement periods Tm as a plurality of feature amounts.
- the extraction circuit 33 may extract each difference between the plurality of extracted first values as a feature amount. You may extract the average value of the difference of as a feature-value.
- the extraction circuit 33 may acquire the third value instead of the first value. That is, the extraction circuit 33 acquires the third value in each of two or more measurement periods Tm among a plurality of consecutive measurement periods Tm-1 to Tm-7 shown in FIGS. A difference between values may be extracted as a feature amount. In this case, the description is made by replacing the first value in the above description with the third value. Further, the extraction circuit 33 may extract both the difference between the first values and the difference between the third values as feature quantities.
- the extraction circuit 33 obtains the difference between the values at the same timing in each of the plurality of measurement periods Tm, thereby obtaining the feature amount corresponding to the drift indicating that the baseline of the signal has fluctuated. can be extracted.
- the extracted feature amount tends to be large.
- the extraction circuit 33 acquires the first value and the second value in the measurement period Tm, and extracts the difference between the acquired first value and the second value as the feature amount.
- the extraction circuit 33 acquires the first value and the second value in at least one measurement period Tm among a plurality of consecutive measurement periods Tm-1 to Tm-7 shown in FIGS. A difference between the first value and the second value obtained from the measurement period Tm is extracted as a feature amount.
- the extraction circuit 33 extracts each of the plurality of differences between the first value and the second value as a feature amount.
- the average value of the differences between the first and second values extracted in plurality may be extracted as the feature amount.
- the extraction circuit 33 can extract the feature amount corresponding to the drift of the signal by taking the difference between the first value and the second value, which is the reference value.
- the extraction circuit 33 acquires the first value in each of two or more measurement periods Tm among a plurality of continuous measurement periods Tm-1 to Tm-7 shown in FIGS. 9 and 10, for example. .
- the extraction circuit 33 acquires the first value in each of all the measurement periods Tm after the second measurement period Tm-2 among a plurality of consecutive measurement periods Tm-1 to Tm-7.
- the extraction circuit 33 may acquire the first value in one or more measurement periods Tm after the second measurement period Tm-2 among a plurality of consecutive measurement periods Tm-1 to Tm-7.
- the extraction circuit 33 extracts all measurement periods Tm after the third measurement period Tm-3 or after the fourth measurement period Tm-4 among a plurality of consecutive measurement periods Tm-1 to Tm-7. A first value in may be obtained.
- the extraction circuit 33 may obtain the first value in each of all the continuous measurement periods Tm-1 to Tm-7.
- the extraction circuit 33 derives an approximation formula using the obtained first value, and extracts the coefficients of the derived approximation formula as feature quantities.
- the approximation formula is, for example, an approximation formula when the first value in each measurement period Tm is a function of time.
- the approximation formula is, for example, a linear formula (that is, linear approximation) or a quadratic formula.
- the approximation formula may be a polynomial formula other than a quadratic formula, and may be an exponential, logarithmic or exponential formula.
- a known method can be used to derive the approximate expression. In the case of a linear approximation, for example, the least squares method can be used to derive the approximation.
- the extraction circuit 33 may acquire the third value instead of the first value. That is, the extraction circuit 33 acquires the third value in each of two or more measurement periods Tm among a plurality of consecutive measurement periods Tm-1 to Tm-7 shown in FIGS.
- An approximation formula may be derived using the values, and the coefficients of the derived approximation formula may be extracted as feature quantities.
- the description is made by replacing the first value in the above description with the third value.
- the extraction circuit 33 may extract both the coefficient of the approximate expression using the first value and the coefficient of the approximate expression using the third value as feature amounts.
- an approximation formula corresponding to the baseline of the signal is derived, so the extraction circuit 33 can extract the feature quantity corresponding to the drift of the signal. Further, by extracting the coefficient of the approximation using the first value or the third value in each of the plurality of measurement periods Tm as the feature quantity, the feature quantity in which the variation in the first value or the third value is smoothed is extracted. can.
- the number of measurement periods Tm for acquiring the first value or the second value may be 3 or more, or 4 or more.
- the extraction circuit 33 may extract at least one feature quantity using any one of the above-described first to third examples.
- a plurality of feature quantities may be extracted using two or more methods.
- the extraction circuit 33 performs the second and subsequent measurement periods Tm (that is, measurement One or more feature quantities may be extracted based on the signal acquired during the period Tm ⁇ 2 or later).
- the feature quantity is extracted using the signal output during the second and subsequent measurement periods Tm in which the drift tends to increase. Therefore, the difference in the extracted feature amount tends to increase according to the type of sample gas, and the identification accuracy described later can be further improved.
- the extraction circuit 33 extracts one or more feature amounts based on the signal acquired during the second and subsequent measurement periods Tm, the extraction circuit 33 does not acquire the first value during the second and subsequent measurement periods Tm. There may be a measurement period Tm.
- the extraction circuit 33 may extract one or more feature amounts based on the signal acquired in any measurement period Tm from the second measurement period Tm onward.
- the extraction circuit 33 may extract one or more feature amounts based on the signals acquired in the third and subsequent measurement periods Tm among the plurality of measurement periods Tm-1 to Tm-7.
- One or more feature quantities may be extracted based on the signal acquired in the fourth and subsequent measurement periods Tm from -1 to Tm-7.
- the extraction circuit 33 may extract a feature amount other than the feature amount corresponding to the drift of the acquired signal.
- the extraction circuit 33 may acquire, as feature amounts, signal values at predetermined intervals in at least one of the second period T2 and the third period T3.
- the predetermined interval is, for example, 0.1 seconds or more and 10 seconds or less.
- the extraction circuit 33 determines the rate of change in the value of the signal in at least one of the second period T2 and the third period T3 and At least one of the amount of change may be acquired as a feature amount.
- 11A and 11B are diagrams for explaining the rate of change (that is, the slope) and the amount of change in the value of the signal output from the sensor 10.
- FIG. 11 shows changes over time in the intensity (for example, voltage) of the signal output from the sensor 10 acquired by the acquisition circuit 32 . In the graph of FIG.
- points a to d are points indicating the time and signal values after a predetermined time has passed since the start of the second period T2
- points e and f are points indicating the third period T2. This point indicates the time and the value of the signal after a predetermined time has elapsed from the start of the period T3.
- the number of points and the position of each point are set according to the type of sensor 10, the type of substance to be identified, and the like. Also, the position of each point may be determined from the shape of the waveform of the signal instead of the predetermined time.
- the extraction circuit 33 extracts, for example, the gradient between two points in the graph of time and signal intensity as a feature quantity.
- the extracted slopes are, for example, the slope SU1 of the line connecting the points a and b in the second period T2, and the line connecting the points c and d in the second period T2 after the point b. and the slope SD of the line connecting the point e and the point f in the third period T3.
- the extraction circuit 33 also extracts, as a feature amount, the amount of change from a predetermined point in time in the graph of signal strength versus time.
- the amount of change extracted is, for example, the amount of change DU, which is the difference between the signal value at the start of the second period T2 and the signal value at point d, and the signal value at the start of the third period T3 and point f is the amount of change DD that is the difference from the value of the signal at .
- the identification circuit 34 uses the learned logical model for identifying the sample gas, based on one or more feature quantities extracted in step S13. , the sample gas is identified, and the identification result is output (step S14).
- the identification circuit 34 uses, for example, a learned logical model, receives one or more feature values as input, and outputs information indicating which of the plurality of identification target substances is contained in the sample gas. .
- the identification circuit 34 may output information indicating whether or not the sample gas contains the identification target substance.
- the trained logical model includes a neural network
- the output node outputs the probability that each of the plurality of identification target substances is contained in the sample gas.
- the learned logic model outputs the identification target substance with the highest probability of being output from the output node among the plurality of identification target substances.
- the learned logic model outputs whether or not the sample gas contains the identification target substance based on the probability output from the output node, for example, based on whether or not the probability is equal to or greater than a threshold.
- the gas identification method performed by the gas identification system 100 is a gas identification method using the sensor 10, and includes an acquisition step (step S12) of acquiring a signal output from the sensor 10, and an acquired signal Using an extraction step (step S13) of extracting one or more feature values corresponding to the drift of the sample gas based on the extracted one or more feature values using the learned logical model, the sample gas is identified and an identification result is output. and an output step (step S14).
- the extraction step the output of the signal during the second period in which the sensor 10 is exposed to the sample gas, that is, the signal drift corresponding to the signal drift, which is different from the feature amount depending on the adsorption concentration of the gas to the sensor 10.
- the above feature values are extracted. Therefore, in the identification step, identification is performed based on signal drift, and sample gases can be identified with high identification accuracy even when sample gases with similar outputs from sensors that depend on gas adsorption concentrations are identified.
- the waveform of the signal in the second period is the same as that of the first sample gas in which the drift of the signal output from the sensor 10 is large, as shown in FIG. , and the second sample gas, in which almost no signal drift occurs, unlike the signal shown in FIG.
- sensors 10 having different materials for the sensing portions were used.
- the sensor 10 in which the material of the sensing portion is made of a resin material such as methylphenyl silicone (75% side chain phenyl groups) or methylphenyl silicone (35% side chain phenyl groups) was used.
- sample gas A phenylethyl alcohol
- sample gas B methylcyclopentenolone
- sample gas C isovaleric acid
- sample gas D undecalactone
- sample gas E skatole
- ⁇ Acquisition of voltage signal> As a voltage signal acquisition operation, first, the 16 sensors 10 are exposed to the reference gas in the first period of 5 seconds, and then the 16 sensors 10 are exposed to the sample gas in the second period of 10 seconds. , followed by exposing the 16 sensors 10 to the reference gas for a third period of 25 seconds. In one operation of acquiring a voltage signal, the exposure operation of the measurement period consisting of the first period, the second period and the third period was continuously repeated seven times. Thus, in one acquisition operation, the 16 sensors 10 were exposed to the sample gas and the reference gas in seven consecutive measurement periods, and voltage signals output from each of the 16 sensors 10 were acquired. .
- a voltage signal set composed of 16 voltage signals corresponding to the 16 sensors 10 was obtained by one signal voltage obtaining operation. Further, such a voltage signal set acquisition operation was performed for the sample gases A to E a total of 158 times. Specifically, 30 voltage signal sets were obtained for sample gas A, and 32 voltage signal sets were obtained for each of sample gases B through E.
- a feature quantity was extracted from each of a total of 158 voltage signal sets corresponding to the sample gases A to E obtained by the above [Acquisition of signal for identification test].
- ⁇ Construction of trained logical model> For each of the training feature sets in the reference example and the working example, 128 sets of training feature sets and identification target substances contained in the sample gas corresponding to the 128 sets of training feature sets are used as teacher data, By performing machine learning using a neural network including one hidden layer with five nodes, a trained logical model for identifying a substance to be identified was constructed. Inputs to the neural network are feature amounts that constitute one set of feature amounts, and outputs from the neural network are probabilities that each of the five substances to be identified is contained in the sample gas. In the learned logical model, the substance to be identified with the highest probability is identified as the substance to be identified contained in the sample gas.
- the learned logical model constructed above was used to identify the identification target substance contained in the sample gas. . Also, the classification of the training feature set and the prediction feature set for the 158 sets was changed, the learned logical model was rebuilt, and the identification target substance contained in the sample gas was identified three times in total.
- Tables 1 to 3 show the results of the first to third identifications using the prediction feature set in the reference example.
- Tables 4 to 6 show the results of the first to third classifications using the prediction feature set in the example.
- the uppermost alphabet and the leftmost alphabet are alphabets corresponding to sample gas A to sample gas E, respectively. Further, in each cell, when inputting a feature amount constituting a prediction feature amount set extracted from signals corresponding to sample gas A to sample gas E listed at the top of the column in which the cell is located, Described is the number of times the learned logic model identifies the substance as the substance to be identified contained in the sample gas described in the leftmost part of the row in which the cell is located. In other words, the number of cells where the sample gas described at the top and the leftmost is the same is the number of times the identification was correct, and the number of cells where the sample gas described at the top and the leftmost is different. is the number of times the identification was wrong.
- the number of mistakes in identifying sample gas D and sample gas E was less than in the reference example.
- the ratio of outputting wrong classification results was 9.4%, 3.1%, and 3.1% for the first to third times, respectively. 1%, and the average of the first to third times was 5.2%.
- the rate of erroneous determination was reduced by 10% or more as compared with the reference example. There was no erroneous determination in the identification of sample gas A to sample gas C.
- the result of the example using the feature amount corresponding to drift has a lower rate of misjudgment than the result of the reference example not using the feature amount corresponding to drift, and the identification accuracy is improved. I know there is.
- the waveform of the signal acquired from the sensor 10 (for example, the sensor 10 whose sensing part is made of methylphenyl silicone)
- the sensor 10 is exposed to the sample gas A and the sensor 10 is exposed to the sample gas B. and the sensor 10 was exposed to the sample gas C, the waveform of the signal in the second period was different. Therefore, in both the reference example and the example, it is considered that the sample gas A to sample gas C were distinguished with high accuracy.
- the exposure unit 20 exposes the sensor 10 to the reference gas in the first period and the third period, but the present invention is not limited to this.
- the exposure unit 20 does not need to expose the sensor 10 to the sample gas during the first period and the third period.
- the sample gas may be sucked to expose the sensor 10 to the vacuum atmosphere. good.
- the gas identification system 100a includes the detection device 200 and the identification device 300, but the present invention is not limited to this.
- the gas identification system 100a may be composed of the identification device 300 only. In this case, for example, the process of step S11 in FIG. 4 is omitted, and the acquisition circuit 32a acquires, for example, already detected sensor signals via the network.
- all or part of the components of the gas identification system according to the present disclosure may be configured with dedicated hardware, or a software program suitable for each component may be executed. It may be realized by Each component may be implemented by a program execution unit such as a CPU or processor reading and executing a software program recorded in a recording medium such as an HDD or a semiconductor memory.
- a program execution unit such as a CPU or processor reading and executing a software program recorded in a recording medium such as an HDD or a semiconductor memory.
- the components of the gas identification system according to the present disclosure may be configured with one or more electronic circuits.
- Each of the one or more electronic circuits may be a general-purpose circuit or a dedicated circuit.
- One or more electronic circuits may include, for example, a semiconductor device, an IC (Integrated Circuit), or an LSI (Large Scale Integration).
- An IC or LSI may be integrated on one chip or may be integrated on a plurality of chips. Although they are called ICs or LSIs here, they may be called system LSIs, VLSIs (Very Large Scale Integration), or ULSIs (Ultra Large Scale Integration) depending on the degree of integration.
- An FPGA Field Programmable Gate Array
- general or specific aspects of the present disclosure may be implemented as a system, apparatus, method, integrated circuit, or computer program. Alternatively, it may be realized by a computer-readable non-temporary recording medium such as an optical disc, HDD, or semiconductor memory storing the computer program. Also, any combination of systems, devices, methods, integrated circuits, computer programs and recording media may be implemented.
- the present disclosure may be implemented as a gas identification method executed by a computer such as a gas identification system, or may be implemented as a program for causing a computer to execute such a gas identification method.
- the present disclosure may be implemented as a computer-readable non-temporary recording medium in which such a program is recorded.
- the gas identification system and gas identification method according to the present disclosure are useful for identifying chemical substances in gas.
- REFERENCE SIGNS LIST 10 sensor 10 exposure unit 21 accommodation unit 22 three-way solenoid valve 23 intake pump 25a, 25b, 25c, 25d, 25e pipes 26a, 26b intake port 26e exhaust port 31 control circuit 32, 32a acquisition circuit 33 extraction circuit 34 identification circuit 40 memory 50 detection unit 51, 60 communication unit 90 wide area communication network 100, 100a gas identification system 200 detection device 300 identification device P1, P2 input port P3 output port
Landscapes
- Chemical & Material Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Combustion & Propulsion (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Electrochemistry (AREA)
- Investigating Or Analyzing Materials By The Use Of Electric Means (AREA)
Abstract
Description
ガスの吸着濃度に応じた信号を出力するセンサをガスの識別に用いる場合、例えば、揮発性有機化合物等の化学物質を含むサンプルガスを暴露した際のセンサから出力される信号を用いた特徴量に基づいて、サンプルガスに含まれる化学物質を識別対象物質として識別する。例えば、化学物質の種類によってセンサへの吸着濃度が違うため、センサから出力される信号が変化する。そのため、例えば、センサをサンプルガスに暴露させる期間におけるセンサから出力される信号の変化量及び変化割合等を特徴量に用いて、サンプルガスに含まれる化学物質の識別が行われる。しかし、化学物質の種類によっては、複数の化学物質の間で、当該期間のセンサへの化学物質の吸着に伴う信号の変化が類似し、識別において誤判定が生じる場合がある。そのため、ガス識別方法では、識別精度を高めることが求められる。 (Circumstances leading to obtaining one aspect of the present disclosure)
When a sensor that outputs a signal corresponding to the adsorption concentration of a gas is used for gas identification, for example, a feature value using the signal output from the sensor when exposed to a sample gas containing chemical substances such as volatile organic compounds , the chemical substance contained in the sample gas is identified as the substance to be identified. For example, the signal output from the sensor changes because the concentration of adsorption to the sensor differs depending on the type of chemical substance. Therefore, chemical substances contained in the sample gas are identified using, for example, the amount of change and the rate of change in the signal output from the sensor during the period in which the sensor is exposed to the sample gas as the feature amount. However, depending on the type of chemical substance, signal changes associated with adsorption of the chemical substance to the sensor during the relevant period are similar among a plurality of chemical substances, and erroneous determination may occur in identification. Therefore, the gas identification method is required to improve the identification accuracy.
本開示の一態様の概要は、以下の通りである。 (Summary of this disclosure)
A summary of one aspect of the disclosure follows.
[構成]
まず、実施の形態に係るガス識別システムの構成について説明する。 (Embodiment)
[Constitution]
First, the configuration of the gas identification system according to the embodiment will be described.
次に、本実施の形態に係るガス識別システムの動作について説明する。以下では、主に、ガス識別システム100の動作について説明するが、特に記載が無い限り、ガス識別システム100aについても同様の動作が行われる。 [motion]
Next, the operation of the gas identification system according to this embodiment will be described. Although the operation of the
図4に示されるように、まず、暴露ステップでは、暴露部20は、測定期間のうち第2期間にのみセンサ10をサンプルガスに暴露させる(ステップS11)。また、暴露部20は、第1期間及び第3期間にセンサ10をリファレンスガスに暴露させる。例えば、制御回路31が、吸気ポンプ23を動作させ、三方向電磁弁22の各ポートの開閉を制御することで、第1期間及び第3期間にセンサ10をリファレンスガスに暴露させ、第2期間にセンサ10をサンプルガスに暴露させる。ステップS11では、例えば、連続する複数の測定期間において、測定期間のうち第2期間にのみセンサ10をサンプルガスに暴露させる。 (1) Exposure Step and Acquisition Step As shown in FIG. 4, first, in the exposure step, the
再び図4を参照し、次に、抽出ステップでは、抽出回路33は、ステップS12で取得された信号のドリフトに対応する1以上の特徴量を抽出する(ステップS13)。抽出回路33は、取得された信号の値を用いて、1以上の特徴量を抽出する。 (2) Extraction Step Referring to FIG. 4 again, in the extraction step, the
再び図4を参照し、次に、識別ステップでは、識別回路34は、サンプルガスを識別する学習済み論理モデルを用い、ステップS13で抽出された1以上の特徴量に基づいて、サンプルガスを識別し、識別結果を出力する(ステップS14)。識別回路34は、例えば、学習済み論理モデルを用いて、1以上の特徴量を入力として、複数の識別対象物質のうちのどの識別対象物質がサンプルガスに含まれているかを示す情報を出力する。識別回路34は、サンプルガスに識別対象物質が含まれているか否かを示す情報を出力してもよい。 (3) Identification Step Again referring to FIG. 4, next, in the identification step, the
次に、本開示を実施例に基づき、具体的に説明する。ただし、本開示は、以下の実施例によって何ら限定されるものではない。 (Example)
Next, the present disclosure will be specifically described based on examples. However, the present disclosure is by no means limited by the following examples.
まず、収容部21に配置された16個のセンサ10を用いて、16個のセンサ10それぞれから出力される信号を取得した。 [Acquisition of signal for identification test]
First, using the 16
16個のセンサには、それぞれ、センシング部の材料の異なるセンサ10を用いた。例えば、センシング部の材料が、メチルフェニルシリコーン(側鎖フェニル基75%)又はメチルフェニルシリコーン(側鎖フェニル基35%)等の樹脂材料で構成されるセンサ10を用いた。 <Sensor>
For the 16 sensors,
リファレンスガスには収容部21が配置された測定室中の空気を用いた。また、サンプルガスには、下記の5種類の化学物質をそれぞれ測定室中の空気に揮発させて得られる5種類のサンプルガスAからEを用いた。つまり、サンプルガスAからサンプルガスEは、それぞれ、下記の対応する化学物質を含む。
・サンプルガスA:フェニルエチルアルコール
・サンプルガスB:メチルシクロペンテノロン
・サンプルガスC:イソ吉草酸
・サンプルガスD:ウンデカラクトン
・サンプルガスE:スカトール <Gas>
As the reference gas, the air in the measurement chamber in which the
・Sample gas A: phenylethyl alcohol ・Sample gas B: methylcyclopentenolone ・Sample gas C: isovaleric acid ・Sample gas D: undecalactone ・Sample gas E: skatole
電圧信号の取得操作としては、まず、5秒間の第1期間において16個のセンサ10をリファレンスガスに暴露させ、続いて、10秒間の第2期間において16個のセンサ10をサンプルガスに暴露させ、続いて、25秒間の第3期間において16個のセンサ10をリファレンスガスに暴露させた。1回の電圧信号の取得操作において、第1期間、第2期間及び第3期間からなる測定期間の暴露操作を連続して7回繰り返した。このように、1回の取得操作では、7回の連続した測定期間において、16個のセンサ10をサンプルガス及びリファレンスガスに暴露させ、16個のセンサ10それぞれから出力される電圧信号を取得した。つまり、1回の信号電圧の取得操作により、16個のセンサ10それぞれに対応する16通りの電圧信号で構成される電圧信号セットを取得した。また、このような電圧信号セットの取得操作を、サンプルガスAからEに対して、合計158回実施した。具体的には、サンプルガスAに対して、30回電圧信号セットを取得し、サンプルガスBからEに対して、32回ずつ電圧信号セットを取得した。 <Acquisition of voltage signal>
As a voltage signal acquisition operation, first, the 16
上記の[識別テスト用信号の取得]によって得られた、サンプルガスAからEに対応する合計158セットの電圧信号セットのそれぞれから、特徴量を抽出した。 [Extraction of feature quantity]
A feature quantity was extracted from each of a total of 158 voltage signal sets corresponding to the sample gases A to E obtained by the above [Acquisition of signal for identification test].
参考例における特徴量の抽出では、電圧信号セットを構成する16通りの電圧信号のそれぞれから、図11に示される傾きSU1、SU2及びSD並びに変化量DU及びDDの5つの特徴量を抽出した。つまり1セットの電圧信号セットに対して、16通り×5つ=80の特徴量で構成される特徴量セットを抽出した。このようにして、158セットの電圧信号セットそれぞれの特徴量セットを抽出し、158セットの参考例における特徴量セットを抽出した。 <Reference example>
In the feature amount extraction in the reference example, five feature amounts of slopes SU1, SU2 and SD and variation amounts DU and DD shown in FIG. 11 were extracted from each of the 16 voltage signals that make up the voltage signal set. In other words, a feature quantity set composed of 16 patterns×5=80 feature quantities was extracted for one voltage signal set. In this way, feature amount sets were extracted for each of the 158 sets of voltage signal sets, and feature amount sets for the reference example of 158 sets were extracted.
実施例における特徴量の抽出では、電圧信号セットを構成する16通りの電圧信号のそれぞれから、参考例と同じ5つの特徴量に加え、信号のドリフトに対応する特徴量である、図9に示される値V1-7と値V1-2との差分、及び、値V3-7と値V3-2との差分の合計7つの特徴量を抽出した。つまり1セットの電圧信号セットに対して、16通り×7つ=112の特徴量で構成される特徴量セットを抽出した。このようにして、158セットの電圧信号セットそれぞれの特徴量セットを抽出し、158セットの実施例における特徴量セットを抽出した。 <Example>
In the extraction of the feature amount in the embodiment, from each of the 16 voltage signals constituting the voltage signal set, in addition to the same five feature amounts as in the reference example, the feature amount corresponding to the signal drift, which is shown in FIG. A total of seven feature amounts were extracted, the difference between the value V1-7 and the value V1-2 and the difference between the value V3-7 and the value V3-2. That is, a feature amount set composed of 16 (16)×7=112 feature amounts is extracted from one voltage signal set. In this way, feature quantity sets were extracted for each of the 158 sets of voltage signal sets, and feature quantity sets for the 158 sets of examples were extracted.
上述の5種類の化学物質を識別対象物質とした場合に、5つの識別対象物質のうちどの識別対象物質がサンプルガスに含まれているかを識別するテストを実施した。 [Identification test]
A test was conducted to identify which of the five identification target substances is contained in the sample gas when the above-described five types of chemical substances are used as the identification target substances.
158セットの参考例及び実施例における特徴量セットを、サンプルガスの種類に関係なく、コンピュータによってランダムに、128セットの参考例及び実施例における訓練用特徴量セットと、32セットの参考例及び実施例における予測用特徴量セットとに振り分けた。なお、上記の特徴量の抽出前に振り分けが実施されてもよい。 <Distribution of feature quantity sets>
158 sets of feature quantity sets in Reference Examples and Examples are randomly generated by a computer regardless of the type of sample gas, 128 sets of feature quantity sets for training in Reference Examples and Examples, and 32 sets of Reference Examples and Examples. It is divided into the prediction feature set in the example. Note that the sorting may be performed before the extraction of the feature amount.
参考例及び実施例それぞれにおける訓練用特徴量セットについての、128セットの訓練用特徴量セットと、128セットの訓練用特徴量セットに対応するサンプルガスに含まれる識別対象物質とを教師データとして、ノード数が5の隠れ層1層を含むニューラルネットワークを用いて機械学習を行うことで、識別対象物質を識別する学習済み論理モデルを構築した。ニューラルネットワークの入力は、1セットの特徴量セットを構成する特徴量であり、ニューラルネットワークの出力は、5つの識別対象物質のそれぞれがサンプルガスに含まれる確率である。学習済み論理モデルでは、最も高い確率の識別対象物質が、サンプルガスに含まれる識別対象物質であると識別される。 <Construction of trained logical model>
For each of the training feature sets in the reference example and the working example, 128 sets of training feature sets and identification target substances contained in the sample gas corresponding to the 128 sets of training feature sets are used as teacher data, By performing machine learning using a neural network including one hidden layer with five nodes, a trained logical model for identifying a substance to be identified was constructed. Inputs to the neural network are feature amounts that constitute one set of feature amounts, and outputs from the neural network are probabilities that each of the five substances to be identified is contained in the sample gas. In the learned logical model, the substance to be identified with the highest probability is identified as the substance to be identified contained in the sample gas.
参考例及び実施例それぞれの32セットの予測用特徴量セットを構成する特徴量を入力として、上記で構築された学習済み論理モデルを用いて、サンプルガスに含まれる識別対象物質の識別を行った。また、158セットに対する訓練用特徴量セットと予測用特徴量セットとの振り分けを変えて、学習済み論理モデルを構築しなおし、サンプルガスに含まれる識別対象物質の識別を合計3回行った。 <Identification of sample gas>
Using the feature values that make up the 32 prediction feature value sets of each of the reference example and the working example as input, the learned logical model constructed above was used to identify the identification target substance contained in the sample gas. . Also, the classification of the training feature set and the prediction feature set for the 158 sets was changed, the learned logical model was rebuilt, and the identification target substance contained in the sample gas was identified three times in total.
以上、本開示に係るガス識別システム及びガス識別方法について、実施の形態及び実施例に基づいて説明したが、本開示は、これらの実施の形態及び実施例に限定されるものではない。本開示の主旨を逸脱しない限り、当業者が思いつく各種変形を実施の形態及び実施例に施したもの、並びに、実施の形態及び実施例における一部の構成要素を組み合わせて構築される別の形態も、本開示の範囲に含まれる。 (Other embodiments)
Although the gas identification system and gas identification method according to the present disclosure have been described above based on the embodiments and examples, the present disclosure is not limited to these embodiments and examples. As long as it does not depart from the gist of the present disclosure, various modifications that a person skilled in the art can think of are applied to the embodiments and examples, and other forms constructed by combining some components in the embodiments and examples are also within the scope of this disclosure.
20 暴露部
21 収容部
22 三方向電磁弁
23 吸気ポンプ
25a、25b、25c、25d、25e 配管
26a、26b 吸気口
26e 排気口
31 制御回路
32、32a 取得回路
33 抽出回路
34 識別回路
40 メモリ
50 検出部
51、60 通信部
90 広域通信ネットワーク
100、100a ガス識別システム
200 検出装置
300 識別装置
P1、P2 入力ポート
P3 出力ポート REFERENCE SIGNS
Claims (13)
- ガスの吸着濃度に応じた信号を出力するセンサを用いたガス識別方法であって、
第1期間、前記第1期間に続く第2期間、及び、前記第2期間に続く第3期間からなる測定期間のうち前記第2期間にのみサンプルガスに暴露させた前記センサから出力される信号を取得する第1ステップと、
取得された前記信号のドリフトに対応する1以上の特徴量を抽出する第2ステップと、
前記サンプルガスを識別する学習済み論理モデルを用い、抽出された前記1以上の特徴量に基づいて、前記サンプルガスを識別し、識別結果を出力する第3ステップと、を含む、
ガス識別方法。 A gas identification method using a sensor that outputs a signal corresponding to the adsorption concentration of a gas,
A signal output from the sensor exposed to the sample gas only during the second period of a measurement period consisting of a first period, a second period following the first period, and a third period following the second period. a first step of obtaining
a second step of extracting one or more features corresponding to the drift of the acquired signal;
a third step of identifying the sample gas based on the extracted one or more feature quantities using a trained logical model for identifying the sample gas, and outputting an identification result;
Gas identification method. - 前記測定期間のうち前記第2期間にのみ前記センサを前記サンプルガスに暴露させる第4ステップをさらに含み、
前記第1ステップでは、前記第4ステップで暴露させた前記センサから出力される前記信号を取得する、
請求項1に記載のガス識別方法。 further comprising a fourth step of exposing the sensor to the sample gas only during the second period of the measurement period;
In the first step, obtaining the signal output from the sensor exposed in the fourth step,
The gas identification method according to claim 1. - 前記第4ステップでは、前記第1期間及び前記第3期間に前記センサをリファレンスガスに暴露させる、
請求項2に記載のガス識別方法。 In the fourth step, the sensor is exposed to a reference gas during the first period and the third period;
The gas identification method according to claim 2. - 前記第1ステップでは、ネットワークを介して前記センサから出力される前記信号を取得する、
請求項1に記載のガス識別方法。 In the first step, the signal output from the sensor is acquired via a network;
The gas identification method according to claim 1. - 前記第2ステップでは、前記第2期間において前記センサが前記サンプルガスに暴露されることによって変動した前記信号の値が前記第3期間において基準値に戻ろうとしている際の前記信号の値である第1値を取得し、取得した前記第1値を用いて前記1以上の特徴量のうちの少なくとも1つの特徴量を抽出する、
請求項1から4のいずれか1項に記載のガス識別方法。 In the second step, the value of the signal fluctuated due to the exposure of the sensor to the sample gas during the second period is the value of the signal when the value of the signal is about to return to the reference value during the third period. Obtaining a first value, and extracting at least one of the one or more feature values using the obtained first value;
The gas identification method according to any one of claims 1 to 4. - 前記第1ステップでは、連続する複数の前記測定期間において前記センサから出力される前記信号を取得し、
前記第2ステップでは、複数の前記測定期間のうちの2以上の前記測定期間それぞれにおいて前記第1値を取得し、取得した前記第1値同士の差分を、前記1以上の特徴量のうちの少なくとも1つの特徴量として抽出する、
請求項5に記載のガス識別方法。 In the first step, the signal output from the sensor is obtained during a plurality of consecutive measurement periods;
In the second step, the first value is obtained in each of the two or more measurement periods among the plurality of measurement periods, and a difference between the obtained first values is calculated as one of the one or more feature quantities. extract as at least one feature,
The gas identification method according to claim 5. - 前記第1ステップでは、連続する複数の前記測定期間において前記センサから出力される前記信号を取得し、
前記第2ステップでは、複数の前記測定期間のうちの2以上の前記測定期間それぞれにおいて前記第1値を取得し、取得した前記第1値を用いて近似式を導出し、導出した前記近似式の係数を、前記1以上の特徴量のうちの少なくとも1つの特徴量として抽出する、
請求項5又は6に記載のガス識別方法。 In the first step, the signal output from the sensor is obtained during a plurality of consecutive measurement periods;
In the second step, the first value is obtained in each of two or more of the plurality of measurement periods, an approximate expression is derived using the obtained first value, and the derived approximate expression The coefficient of is extracted as at least one feature amount of the one or more feature amounts,
The gas identification method according to claim 5 or 6. - 前記第2ステップでは、前記第1期間における最後の前記信号の値である第2値を取得し、取得した前記第1値と取得した前記第2値との差分を、前記1以上の特徴量のうちの少なくとも1つの特徴量として抽出する、
請求項5から7のいずれか1項に記載のガス識別方法。 In the second step, a second value that is the last value of the signal in the first period is obtained, and a difference between the obtained first value and the obtained second value is calculated as the one or more feature quantities. extracted as at least one feature of
The gas identification method according to any one of claims 5 to 7. - 前記第2ステップでは、前記第3期間の最後の前記信号の値を前記第1値として取得する、
請求項5から8のいずれか1項に記載のガス識別方法。 In the second step, the value of the signal at the end of the third period is obtained as the first value;
The gas identification method according to any one of claims 5 to 8. - 前記第1ステップでは、連続する複数の前記測定期間において前記センサから出力される前記信号を取得し、
前記第2ステップでは、複数の前記測定期間のうちの2以上の前記測定期間にそれぞれにおいて、前記第2期間において前記センサが前記サンプルガスに暴露されることによって変動した際の前記信号の値である第3値を取得し、取得した前記第3値同士の差分を、前記1以上の特徴量のうちの少なくとも1つの特徴量として抽出する、
請求項1から9のいずれか1項に記載のガス識別方法。 In the first step, the signal output from the sensor is obtained during a plurality of consecutive measurement periods;
In the second step, in each of the two or more measurement periods among the plurality of measurement periods, the value of the signal when the sensor is exposed to the sample gas during the second period varies. obtaining a certain third value, and extracting a difference between the obtained third values as at least one feature quantity among the one or more feature quantities;
The gas identification method according to any one of claims 1 to 9. - 前記第1ステップでは、連続する複数の前記測定期間において前記センサから出力される前記信号を取得し、
前記第2ステップでは、複数の前記測定期間のうちの2以上の前記測定期間それぞれにおいて、前記第2期間において前記センサが前記サンプルガスに暴露されることによって変動した際の前記信号の値である第3値を取得し、取得した前記第3値を用いて近似式を導出し、導出した前記近似式の係数を、前記1以上の特徴量のうちの少なくとも1つの特徴量として抽出する、
請求項1から10のいずれか1項に記載のガス識別方法。 In the first step, the signal output from the sensor is obtained during a plurality of consecutive measurement periods;
In the second step, in each of two or more of the plurality of measurement periods, the value of the signal when the sensor is exposed to the sample gas during the second period. Obtaining a third value, deriving an approximate expression using the obtained third value, and extracting a coefficient of the derived approximate expression as at least one feature quantity among the one or more feature quantities;
The gas identification method according to any one of claims 1 to 10. - 前記第1ステップでは、連続する複数の前記測定期間において前記センサから出力される前記信号を取得し、
前記第2ステップでは、複数の前記測定期間における2回目以降の前記測定期間で取得した前記信号に基づいて前記1以上の特徴量のうちの少なくとも1つの特徴量を抽出する、
請求項1から11のいずれか1項に記載のガス識別方法。 In the first step, the signal output from the sensor is obtained during a plurality of consecutive measurement periods;
In the second step, extracting at least one of the one or more feature amounts based on the signal acquired in the second and subsequent measurement periods in the plurality of measurement periods,
A gas identification method according to any one of claims 1 to 11. - ガスの吸着濃度に応じた信号を出力するセンサと、
第1期間、前記第1期間に続く第2期間、及び、前記第2期間に続く第3期間からなる測定期間のうち前記第2期間にのみ前記センサをサンプルガスに暴露させる暴露部と、
前記測定期間において前記センサから出力される信号を取得する取得回路と、
取得された前記信号のドリフトに対応する1以上の特徴量を抽出する抽出回路と、
前記サンプルガスを識別する学習済み論理モデルが記憶されるメモリと、
前記学習済み論理モデルを用い、抽出された前記1以上の特徴量に基づいて、前記サンプルガスを識別し、識別結果を出力する識別回路と、を備える、
ガス識別システム。 a sensor that outputs a signal corresponding to the adsorption concentration of the gas;
an exposure unit that exposes the sensor to the sample gas only during the second period of a measurement period consisting of a first period, a second period following the first period, and a third period following the second period;
an acquisition circuit that acquires a signal output from the sensor during the measurement period;
an extraction circuit for extracting one or more feature quantities corresponding to the drift of the acquired signal;
a memory in which a trained logical model identifying the sample gas is stored;
an identification circuit that identifies the sample gas based on the extracted one or more feature values using the learned logical model and outputs an identification result;
Gas identification system.
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202280019745.5A CN116964441A (en) | 2021-03-12 | 2022-03-08 | Gas identification method and gas identification system |
US18/278,802 US20240125748A1 (en) | 2021-03-12 | 2022-03-08 | Gas identification method and gas identification system |
JP2023505569A JPWO2022191173A1 (en) | 2021-03-12 | 2022-03-08 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2021040843 | 2021-03-12 | ||
JP2021-040843 | 2021-03-12 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022191173A1 true WO2022191173A1 (en) | 2022-09-15 |
Family
ID=83227996
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2022/009944 WO2022191173A1 (en) | 2021-03-12 | 2022-03-08 | Gas identification method, and gas identification system |
Country Status (4)
Country | Link |
---|---|
US (1) | US20240125748A1 (en) |
JP (1) | JPWO2022191173A1 (en) |
CN (1) | CN116964441A (en) |
WO (1) | WO2022191173A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024090156A1 (en) * | 2022-10-28 | 2024-05-02 | パナソニックIpマネジメント株式会社 | Smell identification method and smell identification system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0618466A (en) * | 1992-05-07 | 1994-01-25 | Figaro Eng Inc | Gas sensor |
JP2011169830A (en) * | 2010-02-19 | 2011-09-01 | Osaka Prefecture Univ | Odor identification method |
WO2016033186A1 (en) * | 2014-08-26 | 2016-03-03 | Eloret Corporation | Chemical monitoring system |
JP2018105659A (en) * | 2016-12-23 | 2018-07-05 | 国立研究開発法人理化学研究所 | Identification device, identification system, learning device, learning method, identification program, learning program, recording medium, and integrated circuit |
JP2020186934A (en) * | 2019-05-10 | 2020-11-19 | アズビル株式会社 | Capacitive electromagnetic flow meter and measurement control method |
-
2022
- 2022-03-08 WO PCT/JP2022/009944 patent/WO2022191173A1/en active Application Filing
- 2022-03-08 JP JP2023505569A patent/JPWO2022191173A1/ja active Pending
- 2022-03-08 CN CN202280019745.5A patent/CN116964441A/en active Pending
- 2022-03-08 US US18/278,802 patent/US20240125748A1/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0618466A (en) * | 1992-05-07 | 1994-01-25 | Figaro Eng Inc | Gas sensor |
JP2011169830A (en) * | 2010-02-19 | 2011-09-01 | Osaka Prefecture Univ | Odor identification method |
WO2016033186A1 (en) * | 2014-08-26 | 2016-03-03 | Eloret Corporation | Chemical monitoring system |
JP2018105659A (en) * | 2016-12-23 | 2018-07-05 | 国立研究開発法人理化学研究所 | Identification device, identification system, learning device, learning method, identification program, learning program, recording medium, and integrated circuit |
JP2020186934A (en) * | 2019-05-10 | 2020-11-19 | アズビル株式会社 | Capacitive electromagnetic flow meter and measurement control method |
Non-Patent Citations (1)
Title |
---|
LIU QIHE, XIAONAN HU,MAO YE,XIANQIONG CHENG,FAN LI: "Gas Recognition under Sensor Drift by Using Deep Learning", INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS, vol. 30, no. 8, 9 April 2015 (2015-04-09), pages 907 - 922, XP055965926, ISSN: 1098-111X, DOI: 10.1002/int.21731 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024090156A1 (en) * | 2022-10-28 | 2024-05-02 | パナソニックIpマネジメント株式会社 | Smell identification method and smell identification system |
Also Published As
Publication number | Publication date |
---|---|
CN116964441A (en) | 2023-10-27 |
US20240125748A1 (en) | 2024-04-18 |
JPWO2022191173A1 (en) | 2022-09-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Güney et al. | Multiclass classification of n-butanol concentrations with k-nearest neighbor algorithm and support vector machine in an electronic nose | |
El Barbri et al. | Building of a metal oxide gas sensor-based electronic nose to assess the freshness of sardines under cold storage | |
Reimann et al. | Sensor arrays, virtual multisensors, data fusion, and gas sensor data evaluation | |
Mahdavi et al. | Reducing the destructive effect of ambient humidity variations on gas detection capability of a temperature modulated gas sensor by calcium chloride | |
WO2022191173A1 (en) | Gas identification method, and gas identification system | |
CN104569062B (en) | Electronic nose air chamber based on molecular sieve filtration | |
Essiet | Diagnosis of kidney failure by analysis of the concentration of ammonia in exhaled human breath | |
Paknahad et al. | Diffusion-based humidity control membrane for microfluidic-based gas detectors | |
US6360584B1 (en) | Devices for measuring gases with odors | |
CN116457654A (en) | Air quality judging system, air quality judging method and sensor module | |
Wong et al. | Development of a breath detection method based E-nose system for lung cancer identification | |
Bastuck | Improving the performance of gas sensor systems with advanced data evaluation, operation, and calibration methods | |
WO2023037999A1 (en) | Gas analyzing method, and gas analyzing system | |
Brudzewski et al. | An effective method for analysis of dynamic electronic nose responses | |
JP2016186426A (en) | Odor discrimination system | |
Le Maout et al. | A low cost, handheld E-nose for renal diseases early diagnosis | |
Kanade et al. | Development of an E-nose using metal oxide semiconductor sensors for the classification of climacteric fruits | |
Tan et al. | E-nose screening of pesticide residue on chilli and double-checked analysis through different data-recognition algorithms | |
KR101736651B1 (en) | Methods of using information from recovery pulses in electrochemical analyte measurements as well as devices, apparatuses and systems incorporating the same | |
Omatu et al. | Mixed odors classification by neural networks | |
Vergara et al. | Gas sensor drift mitigation using classifier ensembles | |
EP1099949B1 (en) | Device for measuring gases with odors | |
Sharma et al. | Experimental use of electronic nose for odour detection | |
JPH11125610A (en) | Signal-processing device for measuring odor | |
Baumbach et al. | A new method for fast identification of gases and gas mixtures after sensor power up |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22767127 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2023505569 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 18278802 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 202280019745.5 Country of ref document: CN |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 22767127 Country of ref document: EP Kind code of ref document: A1 |